Rabies remains one of the most dreadful infectious diseases affecting human and animals, despite significant scientific advances in its prevention and control. Rabies presents as a distinct problem in different parts of the world. In the Americas, reservoirs of rabies exist in many wild animal species, including raccoons, skunks, foxes, and bats (Rupprecht et al., Emerg. Infect. Dis. 1(4):107-114, 1995). Outbreaks of rabies infections in these terrestrial mammals are found in broad geographic areas across the United States. For example, raccoon rabies affects an area of more than 1 million square kilometers from Florida to Maine. Although wildlife rabies still exists in developed countries, progress has been made in control and elimination of wildlife rabies using oral immunization of wild animals.
Nonetheless, rabies remains a major threat to public health and persists to cause between 50,000 and 60,000 human deaths each year (World Health Organization, April 2003). Humans get infected with the rabies virus mostly through bites from rabid domestic and wildlife animals. In developing countries, dogs are responsible for about 94% of human rabies deaths. Dog rabies is still epizootic in most countries of Africa, Asia and South America and in these countries dogs are responsible for most human deaths from the disease. Controlling rabies virus infection in domestic and wildlife animals, therefore, not only reduces the mortality in these animals but also reduces the risks of human exposure.
The rabies virus is transmitted through broken skin by the bite or scratch of an infected animal. Exposure to rabies virus results in its penetration of peripheral, unmyelineated nerve endings, followed by spreading through retrograde axonal transport, replication occurring exclusively in the neurons, and finally arrival in the central nervous system (CNS). Infection of the CNS causes cellular dysfunction and death (Rupprecht & Dietzschold, Lab Invest. 57:603, 1987). Since rabies virus spreads directly from cell to cell, it largely evades immune recognition (Clark & Prabhakar, Rabies, In: Olson et al., eds., Comparative Pathology of Viral Disease, 2:165, Boca Raton, Fla., CRC Press, 1985).
The rabies virus (RV) is a rhabdovirus—a nonsegmented RNA virus with negative sense polarity. Within the Rhabdoviridae family, rabies virus is the prototype of the Lyssavirus genus. RV is composed of two major structural components: a nucleocapsid or ribonucleoprotein (RNP), and an envelope in the form of a bilayer membrane surrounding the RNP core. The infectious component of all rhabdoviruses is the RNP core, which consists of the negative strand RNA genome encapsidated by nucleoprotein (N) in combination with RNA-dependent RNA-polymerase (L) and phosphoprotein (P). The membrane surrounding the RNP contains two proteins: the trans-membrane glycoprotein (G) and the matrix (M) protein, located at the inner site of the membrane. Thus, the viral genome codes for these five proteins: the three proteins in the RNP (N, L and P), the matrix protein (M), and the glycoprotein (G).
The molecular determinants of pathogenicity of various rabies virus strains have not been fully elucidated. RV pathogenicity was attributed to multigenic events (Yamada et al., Microbiol. Immunol. 50:25-32, 2006). For example, some positions in the RV genome if mutated, affect viral transcription or replication, reducing virulence. Mutations at serine residue 389 of the phosphorylation site in the N gene (Wu et al., J. Virol. 76:4153-4161, 2002) or GDN core sequence of the highly conserved C motif in the L gene (Schnell and Conzelmann, Virol. 214:522-530, 1995) dramatically reduced RV transcription and replication.
The G protein, also referred to as spike protein, is involved in cell attachment and membrane fusion of RV. The amino acid region at position 330 to 340 (referred to as antigenic site III) of the G protein has been identified as important for virulence of certain strains of RV. Several studies support the concept that the pathogenicity of fixed RV strains is determined by the presence of arginine or lysine at amino acid residue 333 of the glycoprotein (Dietzschold et al., Proc. Natl. Acad. Sci. USA 80: 70-74, 1983; Tuffereau et al., Virol. 172: 206-212, 1989).
This phenomenon seems to apply at least to fixed rabies viruses such as CVS, ERA, PV, SAD-B19 and HEP-Flury strains (Anilionis et al., Nature 294:275-278, 1981; Morimoto et al., Virol. 173:465-477, 1989). For example, rabies vaccine viruses possessing an amino acid differing from Arg at position 333 of the glycoprotein are described, for instance, in WO 00/32755 (describing RV mutants in which all three nucleotides in the G protein Arg333 codon are altered compared to the parent virus, such that the Arg at position 333 is substituted with another amino acid); European Patent 350398 (describing an avirulent RV mutant SAG1 derived from the Bern SAD strain of RV, in which the Arg at position 333 of the glycoprotein has been substituted to Ser); and European patent application 583998 (describing an attenuated RV mutant, SAG2, in which the Arg at position 333 in the G protein has been substituted by Glu).
Other strains, such as the RC-HL strain, possess an arginine residue at position 333 of the G, but does not cause lethal infection in adult mice (Ito et al., Microbiol. Immunol. 38:479-482, 1994; Ito et al., J. Virol. 75:9121-9128, 2001). As such, the entire G may contribute to the virulence of RV, although the determinants or regions have not previously been identified. The G gene encodes the only protein that induces viral neutralizing antibody. At least three states of RV glycoprotein are known: the native state (N) being responsible for receptor binding; an active hydrophobic state (A) necessary in the initial step in membrane fusion process (Gaudin, J. Cell Biol. 150:601-612, 2000), and a fusion inactive conformation (I). Correct folding and maturation of the G play important roles for immune recognition. The three potential glycosylated positions in ERA G extracellular domain occur at Asn37, Asn247 and Asn319residues (Wojczyk et al., Glycobiology. 8: 121-130, 1998), respectively. Nonglycosylation of G not only affects conformation, but also inhibits presentation of the protein at the cell surface. Thus, elucidating the molecular determinants underlying pathogenicity of rabies virus presents a complex problem.